CSR: Cascade Conditional Variational Auto Encoder with Socially-aware Regression for Pedestrian Trajectory Prediction

作者：

Highlights：

• The proposed trajectory prediction method consists of a cascaded CVAE module and a socially aware regression module.

• The cascaded CVAE module decouples and balances the loss function with respect to time steps and minimizes the losses at every time steps independently.

• The socially aware regression module corrects the predictions by checking the compatibility between the interaction coding and the crude predicted trajectories.

摘要

•The proposed trajectory prediction method consists of a cascaded CVAE module and a socially aware regression module.•The cascaded CVAE module decouples and balances the loss function with respect to time steps and minimizes the losses at every time steps independently.•The socially aware regression module corrects the predictions by checking the compatibility between the interaction coding and the crude predicted trajectories.

论文关键词：Pedestrian trajectory prediction,Socially-aware model,Conditional variational autoencoder (CVAE)

论文评审过程：Received 18 November 2021, Revised 27 July 2022, Accepted 4 September 2022, Available online 7 September 2022, Version of Record 15 September 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.109030